Side Effects: A Prosecutor, a Whistleblower, and a Bestselling Antidepressant on Trial

by Alison Bass

  In April 1998, McLean and Sepracor completed the contract, which called for McLean to receive $1 million up front and then 10 percent from sales of the new drug. The Harvard teaching hospital collected 50 percent of the up-front money, or $500,000; Teicher's lab received 25 percent, or $250,000, according to the terms of the contract; and Teicher, as one of the co-inventors of R-, walked away with $250,000 for himself. He invested most of it in the stock market, which was soaring on the wings of an innovation known as the Internet.

  LITTLE MORE THAN a month later, Marty Teicher found himself staring out the window of a shuttle bus that was taking him from his hotel to Toronto's sleek-looking convention center several blocks away. He had come to Toronto to attend two conferences: the annual meetings of the American Psychiatric Association (APA) and the Society of Biological Psychiatry. This year, the two back-to-back meetings had drawn more than twenty thousand psychiatrists and psychiatric researchers to a bustling four-block destination in Canada's largest metropolis.

  Teicher, who saw himself foremost as a researcher with a few private patients on the side, liked to spend much of his time at the biological psychiatry conference because most of its sessions were devoted to new research. By contrast, the APA meeting was intended for the general practitioner-the non-research-oriented psychiatrist who needed to catch up on new developments in clinical practice. The APA relegated its new research findings mostly to poster sessions, in which an abstract and some charts were slapped on a poster and hung in a large exhibit room. The lead author usually made a brief scheduled appearance to discuss his results with whoever happened by. Teicher found the poster sessions less satisfying than the research presentations at the biological society and the lively question-and-answer sessions that followed. Moreover, in recent years, he had noticed that the pharmaceutical industry dominated the APA conference, bankrolling huge, well-attended symposia in the convention's biggest ballrooms. Many of these industry-supported symposia pretended to be about new developments in one mental illness or another, but to Teicher's way of thinking, they were really about promoting the drugmakers' newest products.

  Just the previous day, he himself had attended an industry-supported symposium in the Grand Ballroom of the Sheraton Centre Hotel titled "Chronic Depression: Optimizing Long-Term Treatment." Teicher had decided to sit in on the three-hour session because it was being chaired by an old friend of his, Dr. Alan Schatzberg, a prominent psychiatrist who had mentored him at McLean. Schatzberg was now head of the Massachusetts Mental Health Center in Boston but still kept a research lab at McLean.

  It just so happened that the symposium chaired by Schatzberg was being funded by Bristol-Myers Squibb (BMS), a pharmaceutical company with a hot new antidepressant on the market called Serzone. Unbeknownst to Teicher and the hundreds of other doctors in attendance, all three of the speakers at the symposium moderated by Schatzberg were principal investigators in a newly finished study of Serzone, also funded by Bristol-Myers Squibb. The study, which concluded that Serzone was spectacularly effective in treating chronic depression when prescribed in conjunction with psychotherapy, would be published with much fanfare in the New England Journal of Medicine two years later, in May 2000. Its publication would precipitate an unusual editorial by the journal's editor, titled "Is Academic Medicine for Sale?" In her editorial, then-editor Marcia Angell would decry the fact that all but one of the twelve principal authors of the Serzone study had "extensive financial associations" with Bristol-Myers Squibb and other pharmaceutical companies. Indeed, the list of their financial ties to the drug industry was so long that the journal could not fit the full text in its print version of the article. (In 2003, a number of countries would ban the sale of Serzone after it caused liver failure and the deaths of at least twenty patients. Bristol-Myers Squibb would pull its once-promising antidepressant from the U.S. market in May 2004.)

  At the 1998 APA meeting in Toronto, however, the buzz about Serzone was all positive. The first speaker at the Bristol-Myers Squibb symposium that morning was none other than Dr. Martin Keller, chief of psychiatry at Brown University. As it turned out, Keller was also the principal investigator of the newly finished Serzone study, but no mention of that was made to the five-hundred-odd doctors attending his talk, which focused on the clinical course of chronic depression. Nor did Keller disclose that he was not there on his own dime. According to testimony in a later lawsuit, Brown's psychiatry chief had been paid $1,500, plus $2,000 in travel expenses, to be one of the featured speakers that day. The other speakers, along with the moderator, were similarly well reimbursed for their time.

  Teicher had never actually met Keller, though he had been one seat away from the Brown psychiatrist on a plane heading toward San Juan for the annual meeting of the American College for Neuropsychiatry (ACNP) the previous December. Engrossed in sorting through his slides, Teicher had not paid much attention to the woman sitting next to him or to the elegantly dressed man in the aisle seat. Then the couple, who Teicher felt were obviously married, began arguing, and he couldn't help looking up from his slides. Only later, when he saw Martin Keller give a presentation at the ACNP meeting, did he realize who his flying companions had been.

  Keller, he knew, was a very successful guy. Mutual colleagues had described him as smart, funny, and well connected to the circle of psychiatrists appointed to top positions in the National Institute of Mental Health, the nation's premier funding agency for mental health research. Keller indeed seemed to have a golden touch when it came to winning research grants from the NIMH.

  Keller was also a sought-after speaker on the psychiatry circuit, and his schedule for the 1998 APA conference in Toronto reflected this status. According to the APA program, he was participating that year in two industry-supported symposia. One was the Bristol-Myers Squibb symposium that Schatzberg chaired; the other was sponsored by Pfizer, on the topic of managing depression among baby boomers. Keller himself was chairing that session. And as it turned out, he had just put the finishing touches on a paper for JAMA that contained great news about Zoloft, Pfizer's SSRI product. The JAMA paper, which would be published in a few months' time with Keller as the lead author, concluded that Zoloft was significantly more effective than placebo in preventing the recurrence of major depression in chronically depressed patients.

  Keller was also on the APA program in 1998 as the lead presenter of a poster session on "Paroxetine and Imipramine in the Treatment of Adolescent Depression." This was the newly finished multicenter trial sponsored by SmithKline Beecham that Keller had been principal investigator of, it compared paroxetine (Paxil) to placebo and an older antidepressant known as imipramine. According to the program's abstract, the study's results showed that "paroxetine is an effective treatment for major depression in an adolescent outpatient population." It was the first public announcement of a study that would receive much scrutiny in the coming years.

  IT SEEMED TO TAKE Teicher's shuttle bus forever to travel the few blocks to the convention center. He should have walked. The bus finally pulled up to the curb. Teicher hopped out and saw Alan Schatzberg in the group milling about outside the entrance. Schatzberg was some years older than Teicher, a gregarious, backslapping type of guy whom Marty fondly referred to as "one of the world's greatest schmoozers" They shook hands, and Schatzberg clapped Teicher on the back.

  "It's great to see you, Marty," Schatzberg said. "Hey, do you have time for a quick walk? It's such a nice day!"

  As the two men ambled along Lakeshore Boulevard, enjoying the crisp late spring breeze off Lake Ontario, Schatzberg wanted to know all about what Teicher was working on, how things were going at McLean. Teicher was flattered by the older man's attention.

  "Things are going pretty well," Teicher replied. He was working on some promising research about the long-term consequences of childhood abuse. His lab, in fact, was in the midst of a five-year, $1.5 million grant from the National Institute of Mental Health to study the impact of childhood abuse on changes in the brain. T
eicher knew that a few of his colleagues privately suspected that his pummeling by Lilly both inside and outside the courtroom-may have been one reason why he had shifted his research focus away from Prozac. But Teicher later offered a different explanation for his change in focus: "Once I became involved in this patent for Sepracor and felt there was potential for profiting from it [an alternative for Prozac], I felt it would be a conflict to continue studying and speaking out about Prozac," he said.

  During their stroll around downtown Toronto, Teicher did not get into any of this with his friend. And Schatzberg did not mention the symposium on chronic depression he had chaired, or its speakers' close ties to Bristol-Myers Squibb. Looking back, Teicher doubted that Schatzberg even knew the extent to which Martin Keller and some of the other speakers that day were profiting from their involvement with pharmaceutical companies.

  It would be another sixteen months before Keller's colleagues, along with the general public, learned the full extent of Keller's pharmaceutical winnings. In a front-page story for the Boston Globe on October 4, 1999, I reported that the Brown psychiatry chief was earning hundreds of thousands of dollars (above and beyond his salary from Brown) in personal consulting and speaking fees from the pharmaceutical industry.

  Even before the Globe published that story, though, it had been no secret in psychiatric circles that Keller was a friend of Big Pharma. His six-page resume from the mid-1990s openly lists him as a consultant to seven of the largest pharmaceutical companies then in existence: Upjohn, Bristol-Myers Squibb, Eli Lilly, Johnson and Johnson, Pfizer, Sandoz, and SmithKline Beecham. And on the disclosure form for the 1998 APA meeting, in which participants were required to note "any significant financial interests or affiliations" that might conflict with commercial products discussed in the scientific program, Keller's list included just about every drug company then marketing antidepressants in the United States: Wyeth-Ayerst (Effexor), SmithKline Beecham (Paxil), Pfizer (Zoloft), Pharmacia and Upjohn (Vestra), Bristol-Myers Squibb (Serzone), and Eli Lilly (Prozac).

  Even so, some of his colleagues were startled to discover just how extensive the Brown chief's conflicts of interest were. Two of the drug companies providing Keller with especially generous stipends in 1998Pfizer (which paid him $218,000) and Bristol-Myers Squibb (which gave him $77,400)-stood to benefit handsomely from the positive findings that Keller and his colleagues would soon publish (about Zoloft and Serzone) in prestigious medical journals.

  In 1998, according to his own tax returns, Keller also earned $62,500 in personal income from Forest Labs, the distributor of Celexa, a relatively new entry to the SSRI market; $19,669 from Merck, the maker of Roxindole, a promising new antidepressant at the time; $8,785 from Wyeth-Ayerst, producer of the antidepressant Effexor; and $8,625 from Organon, which made Remeron, yet another antidepressant just coming onto the market. And that was only a partial list of the money he collected from the pharmaceutical industry. According to his tax returns Keller pulled in a total of $444,000 in 1997 and $556,000 in 1998 from firms either making drugs or promoting them.

  Keller declined to comment for the Boston Globe article in 1999. But several ethicists I quoted in the article said that such unusually large consulting fees constituted the most serious potential conflict they'd ever heard of in academic medicine. They also expressed dismay that Keller had not fully disclosed his financial conflicts to the medical journals that published his research, or to the American Psychiatric Association.

  "If I were this person's supervisor, I would have major concerns that such a large sum of money would bias his thinking and judgment," said Dr. Arnold Relman, a former editor of the New England Journal of Medicine and professor emeritus of medicine at Harvard Medical School.

  In the 1990s, academic medicine was just beginning to sort through the ethical conundrums posed by the enormous sums companies were pouring into drug development and testing. Until the advent of managed care in the late 1980s, drugmakers had been able to raise their prices at will and were not under much pressure to bring new products to market. But when managed care plans began to squeeze their profits, drug companies felt compelled to increase the number of drugs they sold and speed up development times. Faculty at many medical schools joined the stampede, finding clinical trials to be a particularly lucrative source of income. According to a survey conducted by the New York Times in 1999, drugmakers were paying doctors between $1,600 and $4,581 per patient for clinical trials. Between 1990 and 1998, according to his resume and his department's annual report, Martin Keller brought in nearly $ 8.7 million in research funding from pharmaceutical companies, many of which were also paying him thousands of dollars in personal consulting income.

  By then, a few of the more prestigious medical journals, such as the New England Journal of Medicine and the Journal of the American Medical Association, had begun requiring that authors disclose if they had "substantial" financial ties (more than $10,000) to companies whose products they were studying-either from consulting income or from equity in the companies. But none of the journals required authors to reveal the exact amount of their earnings. And while a handful of top medical schools, such as Harvard and Georgetown, began actively discouraging faculty from doing research for companies who were paying them more than $10,000 or $20,000 a year in consulting fees, most universities merely required faculty to complete pro forma disclosure forms, again without revealing the exact dollars going into their bank accounts. Federal funding agencies like NIH and NIMH had similar requirements, as did professional organizations like the American Psychiatric Association. But these policies were rarely enforced. When my Boston Globe article revealed that Martin Keller had failed to fully disclose the extent of his financial ties to the drug industry, Brown University leaped to their star rainmaker's defense, saying it had no problem with Keller's consulting arrangements. Likewise, top officials at the APA and at NIMH, which had given Keller millions of dollars in research grants over the years, took no punitive action against him. "Alas, knowing there is impropriety and proving there is impropriety are vastly different issues," said Dr. Stephen Goldfinger, chair of the Department of Psychiatry at SUNY Downstate Medical School and chair of the APA Committee on Commercial Support.

  As a scientist himself, Teicher had no problem with academics' making money from research. After all, he stood to benefit handsomely from the contract he had signed with Sepracor for inventing R-. Teicher, in fact, would make out very well when Lilly bought the rights to R- from Sepracor for $20 million in December 1998. The Indianapolis-based company was hoping to have the R- ready for market as an antidepressant by the time its patent for Prozac expired in 2001. Of the $20 million the drug company paid to Sepracor for the rights to R-, $2 million went directly to McLean Hospital, with Teicher's lab and Teicher himself each netting $500,000.

  Despite this windfall, Teicher held a rather conservative view of research conflicts. Like many of his colleagues at Harvard, he took the medical school's strict conflict-of-interest policy to heart. Years later he would note that there is a distinct difference between getting royalties for a drug one helped develop and earning thousands of dollars in personal income from the very companies whose drugs you were touting in medical journals and at conferences.

  "When you do that, you've become a shill for the company," Teicher said. "You want to continue getting those payments, so you're biased toward saying favorable things about their drug product. By contrast, most institutions consider getting royalties from a company that has licensed your work as not posing a conflict, as long as you're not sitting on its board of directors and getting consulting income from the company."

  Teicher is far from alone in drawing that distinction. In 2004, researchers at the Social Policy Research Institute in Illinois would publish a study showing that authors with self-admitted conflicts of interest were ten to twenty times less likely to present negative findings than those without such financial conflicts. This inverse relationship was strongest among researchers stu
dying drug treatments. In their study, published in the Journal of General Internal Medicine, the authors concluded, "Conflict of interest is widespread among the authors of published manuscripts and these authors are more likely to present positive findings."

  By then, Sheldon Krimsky, a Tufts University professor and author of the book Science in the Private Interest, had also found that the amount of money consultants receive makes a difference. "If you're only get ting a thousand dollars for a year, you would have a different response than if you're getting two hundred thousand a year," Krimsky said. "At some point the quantity of money produces a qualitative change in people, and the integrity of the scientific endeavor is tarnished."

  But in 1998, such findings had yet to be drawn, and the overweening influence of the pharmaceutical industry on medical research was not part of the public debate. The notion that science was pure-that money and prestige couldn't possibly influence a scientist's judgment-held sway. Even journalists, by and large, believed this party line. In reporting on the latest health findings, most medical writers took what they read in press releases and journals at face value and did not dig much further.